The OpenClaw ecosystem shipped v2026.6.9 this week with richer Telegram delivery, deeper Codex integration, and security patches against vCard/location injection attacks — while simultaneously facing a reckoning as 1,184 malicious skills were confirmed on ClawHub and 492 MCP servers were found exposed with zero authentication. Anthropic raised eyebrows with a leaked "Conway" always-on agent, even as its hiring spree landed DeepMind Nobel laureate John Jumper alongside Karpathy and Ross Nordeen. On the payments front, Visa and Mastercard concurrently launched agentic payment rails, while the Five Eyes alliance formally urged "careful adoption" of agentic AI. Elon Musk predicted AI compute will move to space within 36 months, and NVIDIA deepened its Hyundai partnership for humanoid robots — the physical AI era is accelerating alongside the software one.
OpenClaw's June release adds structured text rendering on Telegram (tables, lists, expandable blockquotes), deeper Codex integration, safer model routing for GLM-5.2 and Claude Haiku 4.5, and critical patches hardening Docker sandbox boundaries and browser CDP endpoint validation.
Antiy CERT confirmed 1,184 malicious skills across ClawHub, the marketplace for the OpenClaw AI agent framework. Trend Micro found 492 MCP servers exposed to the internet with zero authentication. Check Point disclosed remote code execution in Claude Code through poisoned repo configs. The Pentagon designated Anthropic a "supply chain risk" — the first American company to receive the classification.
Flowtivity's updated comparison rates security, ease of use, and ecosystem maturity. Key finding: OpenClaw's sandboxed execution and command approval system provide stronger guardrails, while Hermes Agent wins on non-coder accessibility and simpler setup.
devFlokers' monthly roundup highlights a significant shift toward technological accessibility: efficient sparse models, local-first system integration tools like TencentDB-Agent-Memory (5,500 GitHub stars), and dedicated desktop AI accelerators are decentralizing AI beyond cloud APIs.
IT Business Today examines the tension at the heart of the agentic AI wave: OpenClaw has made self-hosted AI agents genuinely useful for the first time, but the speed of adoption is outpacing enterprise security readiness. The first wave was chatbots; this wave is autonomous action — with real consequences.
Selected posts from the edition, with links to the originals.
"If you look at who Anthropic has hired in the last 6 months, it's actually WILD. Read this roster twice: Andrej Karpathy (OpenAI/Tesla, May 2026), John Jumper (Google DeepMind, June 2026), Ross Nordeen (xAI/Tesla, May)"
View post on X ↗"Don't sleep on Codex. I've been using it since day one. GPT-5.5 xHigh + full access is beast mode. Computer use + browser use + spawned threads make agent loops incredibly powerful. Picking up tasks from mobile works surprisingly well."
View post on X ↗"Anthropic's always-on 'Conway' agent leaked from their codebase, OpenClaw v2026.6.9 ships CI gating and fail-closed approvals, plus a new AI coding agent how-to."
View post on X ↗"OpenClaw v2026.6.5 drops with Codex improvements — the agent framework gets steadier recovery, richer Telegram delivery, and slimmer distribution. Not flashy, but the Codex improvements alone make this worth updating."
View post on X ↗"I met a guy last night building a next-level voice harness. He hooked Claude up to it and it realized quickly the way the voice model worked and optimized itself for that use."
View post on X ↗"When even @Vercel's CEO says this about GLM 5.2, people should probably take notice 👀"
View post on X ↗"Elon Musk reveals why he believes the cheapest place to put AI will be space within 36 months. 'The availability of energy is the issue. Everywhere outside of China, electrical output is more or less flat. The output of chips is growing exponentially, but the output of energy is not.'"
View post on X ↗"People asking me: Claude Cowork or OpenClaw for a Mac AI agent? Cowork is an appliance. OpenClaw is a workshop. One gets you productive in 5 minutes. The other lets you build anything."
View post on X ↗"June 10: Visa & Mastercard launched agentic payments same day. Visa + OpenAI. Mastercard Agent Pay. Europe ran live AI agent payments. Same week: Alipay launched Ah Bao — an AI agent for 1B users. The US-China AI war just found a new front: payments."
View post on X ↗"Hermes Didn't Win, OpenClaw Faded" — a provocative deep dive on the AI agent platform competition, arguing that the real winner may be neither of today's frameworks but the ecosystem infrastructure being built around them."
View post on X ↗